BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160904Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess322_post219@linklings.com
SUMMARY:Automatic Generation of Mixed-Precision Programs
DESCRIPTION:Poster\nTech Program Reg Pass, Exhibits Reg Pass\n\nAutomatic 
 Generation of Mixed-Precision Programs\n\nMoody, Pinnow, Lam, Menon, Schor
 dan...\n\nFloating-point arithmetic is foundational to scientific computin
 g in HPC, and choices about floating-point precision can have a significan
 t effect on the accuracy and speed of HPC codes. Unfortunately, current pr
 ecision optimization tools require significant user interaction, and few w
 ork on the scale of HPC codes due to significant analysis overhead. We pro
 pose an automatic search and replacement system that finds the maximum spe
 edup using mixed precision given a required level of accuracy. To achieve 
 this, we integrated three existing analysis tools into a system that requi
 res minimal input from the user. If a speedup is found, our system can pro
 vide a ready-to-compile mixed-precision version of the original program.
URL:https://sc18.supercomputing.org/presentation/?id=post219&sess=sess322
END:VEVENT
END:VCALENDAR